Method and system for estimating asset location within a material handling environment

Abstract

Various embodiments disclose a method for tracking an asset. The method includes determining one or more locations of the asset within an indoor environment based on metadata associated with RF signals received from an RF tag associated with the asset. The method further includes identifying a first set of locations of the one or more locations. The method further includes identifying a second set of locations of the one or more locations, wherein the second set of locations correspond to uncalibrated locations of the asset within the indoor environment. Additionally, the method includes receiving a third set of locations of the asset. Furthermore, the method includes training a machine learning (ML) model based on the first set of locations, the second set of locations, and the third set of locations, and the metadata associated with the RF signals. The ML model predicts a fourth set of locations of another asset.

Description

Technical Field

[0001] The exemplary solutions disclosed herein relate generally to a material handling environment, and more specifically, to a method and system for estimating the location of assets within the material handling environment. Background Technology

[0002] A typical material handling environment includes one or more shelves, which can be configured to store assets such as, but not limited to, cartons, boxes, packages, etc. In some examples, it may be necessary to track the location of these assets within the material handling environment. For this purpose, the material handling environment may be equipped with multiple radio frequency (RF) beacons located at predetermined locations within the material handling environment. These multiple RF beacons can be configured to determine location based on signals received from RF tags coupled to each asset. Summary of the Invention

[0003] The various embodiments shown herein disclose a method for tracking assets. The method includes determining one or more locations of the asset within an indoor environment based on metadata associated with RF signals received from an RF tag associated with the asset. The method further includes identifying a first set of locations among the one or more locations. The first set of locations corresponds to a calibrated location of the asset within the indoor environment. The method further includes identifying a second set of locations among the one or more locations, wherein the second set of locations corresponds to an uncalibrated location of the asset within the indoor environment. Additionally, the method includes receiving a third set of locations for the second set of locations of the asset, wherein the third set of locations corresponds to a calibrated location of the second set of locations. Furthermore, the method includes training a machine learning (ML) model based on the first set of locations, the second set of locations, and the third set of locations, as well as the metadata associated with the RF signals, wherein the ML model is configured to predict a fourth set of locations for another asset within the indoor environment.

[0004] The various embodiments shown herein disclose a central server for tracking assets. The central server includes a processor. Furthermore, the central server includes a memory device communicatively coupled to the processor, the memory device including a set of instructions executable by the processor to determine one or more locations of the asset within an indoor environment based on metadata associated with radio frequency (RF) signals received from an RF tag associated with the asset. The processor is further configured to identify a first set of locations among the one or more locations, wherein the first set of locations corresponds to locations within the indoor environment where the RF tag on the asset is within the line of sight (LOS) of an RF beacon installed in the indoor environment. Additionally, the processor is configured to identify a second set of locations among the one or more locations, wherein the second set of locations corresponds to locations within the indoor environment where the RF tag is outside the LOS of the RF beacon, wherein the second set of locations corresponds to an uncalibrated location of the RF tag within the indoor environment. Furthermore, the processor is configured to receive a third set of locations, wherein the third set of locations corresponds to calibrated locations of the second set of locations. The processor is further configured to train an ML model based on a first set of locations, a second set of locations, a third set of locations, and metadata associated with a machine learning (ML) model, wherein the ML model is configured to predict a fourth set of locations of another asset within an indoor environment when the RF tag on the other asset is outside the LOS of the RF beacon.

[0005] The various embodiments shown herein disclose a method for tracking assets. The method includes determining one or more locations of an asset within an indoor environment based on metadata associated with radio frequency (RF) signals received from an RF tag associated with the asset, wherein the one or more locations include a first set of locations and a second set of locations, wherein the first set of locations corresponds to a calibrated location of the asset within the indoor environment, and wherein the second set of locations corresponds to an uncalibrated location of the asset within the indoor environment. Furthermore, the method includes having a processor predict a third set of locations based on the first and second sets of locations using a machine learning (ML) model. Additionally, the method includes predicting the location of a channel for the stored asset based on the third set of locations.

[0006] The above-described exemplary invention, as well as other exemplary objects and / or advantages of this disclosure and the ways in which these objects and / or advantages can be achieved, can be further explained in the following detailed description and accompanying drawings. Attached Figure Description

[0007] The description of the exemplary embodiments can be read in conjunction with the accompanying drawings. It should be understood that, for simplicity and clarity of illustration, the elements shown in the drawings are not necessarily drawn to scale. For example, the dimensions of some elements are exaggerated relative to others. Embodiments incorporating the teachings of this disclosure are shown and described with reference to the accompanying drawings presented herein, in which:

[0008] Figure 1An exemplary material handling environment according to one or more embodiments described herein is shown;

[0009] Figure 2 A block diagram of a central server according to one or more embodiments described herein is shown;

[0010] Figure 3 A flowchart is shown of a method for operating a central server according to one or more embodiments described herein;

[0011] Figure 4 A flowchart is shown of a method for training another ML model capable of predicting the accurate location of assets stored in shelves and / or aisles, according to one or more embodiments described herein;

[0012] Figure 5 A flowchart is shown of a method for predicting a fourth set of locations of another asset within a material handling environment, according to one or more embodiments described herein.

[0013] Figure 6 A flowchart is shown of a method for classifying time-series data of location according to one or more embodiments described herein;

[0014] Figure 7 A flowchart is shown of a method for classifying time-series data of location according to one or more embodiments described herein;

[0015] Figure 8 A flowchart is shown of a method for classifying a set of crossing locations into a first group of locations and a second group of locations according to one or more embodiments described herein;

[0016] Figure 9 A flowchart is shown illustrating a method for determining a third set of positions for a second set of positions, according to one or more embodiments described herein; and

[0017] Figure 10 An exemplary scenario for training an ML model according to one or more implementations described herein is shown. Detailed Implementation

[0018] Some embodiments of this disclosure will be described more fully below with reference to the accompanying drawings, which illustrate some, but not all, embodiments of this disclosure. In fact, this disclosure may be embodied in many different forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided to enable this disclosure to meet applicable legal requirements. Throughout this document, similar reference numerals refer to similar elements.

[0019] Unless the context otherwise requires, throughout this specification and the following claims, the word “comprising” and its variations such as “including” and “having” shall be interpreted in an open sense, that is, as “including but not limited to”.

[0020] Throughout this specification, references to “an embodiment” or “an embodiment” mean that a particular feature, structure, or characteristic described in connection with that embodiment is included in at least one embodiment. Therefore, the appearance of the phrase “in an embodiment” or “in an embodiment” in various places throughout this specification does not necessarily refer to the same embodiment. Furthermore, one or more particular features, structures, or characteristics from one or more embodiments may be combined in any suitable manner in one or more other embodiments.

[0021] As used herein, the terms “example” or “exemplary” mean “serving as an example, instance, or illustration.” Any specific implementation described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other specific implementations.

[0022] If the specification states that a component or feature "may," "can," "should," "will," "preferably," "possibly," "usually," "optionally," "for example," "often," or "may" (or other such language) be included or have that characteristic, then the specific component or feature is not necessarily included or has that characteristic. Such components or features may be optionally included in some embodiments or excluded.

[0023] The terms “electronically coupled,” “electronically coupling,” “electronically couple,” “communicating with,” “electronically communicating with,” “communicationally coupled,” or “connection” in this disclosure mean that two or more components are connected (directly or indirectly) via wired means (e.g., but not limited to system bus, wired Ethernet) and / or wireless means (e.g., but not limited to Wi-Fi, Bluetooth, ZigBee) such that data and / or information can be sent to and / or received from these components.

[0024] As used herein, the term "antenna element" refers to a device or apparatus (e.g., an active element) that can be configured to generate a radio frequency (RF) signal when a voltage signal is applied to the antenna element. For example, an antenna element may be configured to generate an RF signal in a high frequency (HF) band or a low frequency (LF) band. Alternatively, an antenna element may generate an RF signal in an ultra-high frequency (UHF) band. Alternatively, an antenna element may generate an RF signal in other frequency bands. In some examples, the antenna element may further include matching circuitry, which is, for example, electronically coupled to the active element to generate the RF signal.

[0025] The term "radio frequency (RF) tag" is used herein to refer to an article of manufacture, device, or apparatus that may include an integrated circuit (IC), an antenna element, and a substrate. In exemplary embodiments, the antenna element and IC may be fabricated on a substrate. In exemplary embodiments, the IC may be communicatively coupled to the antenna element via interconnects on the substrate. In exemplary embodiments, the integrated circuit in the RF tag may be configured to store encoded information or encoded data. In some examples, the RF tag may be configured to operate in one or more RF bands, such as, but not limited to, 3 MHz to 30 MHz (HF band), 2.4 GHz, 5 GHz, and / or 860 MHz to 960 MHz (UHF band). In some exemplary embodiments, the RF tag may have a dedicated power supply that enables the RF tag to communicate with one or more components. Such RF tags are referred to as active RF tags. In alternative exemplary embodiments, the RF tag may not have a dedicated power supply. Such RF tags are referred to as passive RF tags. In such embodiments, the RF tag may have a power coupler capable of inducing a charge when the RF tag is brought into an RF field. The induced charge can then be used to power the RF tag itself.

[0026] An RF system may include one or more RF beacons. These RF beacons may be configured to continuously or periodically receive RF signals from multiple RF tags (placed on one or more assets). Furthermore, the RF beacons may be configured to periodically transmit data received from the one or more RF tags to a central server. To facilitate data transmission to the central server, the one or more RF beacons may be communicatively coupled to the central server via a backbone network, such as, but not limited to, wireless networks, Ethernet networks, etc.

[0027] The term "material handling environment" may correspond to a predefined area that facilitates operations such as loading / unloading and storing finished goods. Alternatively, a material handling environment may involve storing finished goods in one or more shelving / aisles. Some examples of material handling environments may include, but are not limited to, warehouses, retail outlets, etc.

[0028] Typically, material handling systems involve storing assets in predetermined locations, such as in one or more aisles and / or one or more shelves. One or more RF tags coupled to each asset can be used to track the location of the assets within the material handling environment. For example, the RF tags can be configured to periodically send beacon signals to one or more RF beacons installed at predetermined locations within the material handling environment. For example, the RF tags can be configured to broadcast beacon signals. In an exemplary embodiment, a group of one or more RF beacons can receive beacon signals from the RF tag. In some examples, a group of RF beacons may be located within the line of sight (LOS) of the RF tag and therefore can receive beacon signals from the RF tag. In another example, a group of RF beacons may receive the beacon signal after it has likely experienced one or more reflections (from one or more surfaces in the material handling environment). Such reflections of the beacon signal can occur when the asset approaches one or more surfaces. For example, such reflections of the beacon signal can occur when the asset is near an aisle and / or shelf and is about to be stored in the aisle. In another example, such reflections of the beacon signal can occur when the asset is positioned in an aisle and / or shelf.

[0029] In some examples, a set of RF beacons may further transmit beacon signals to a central server. In exemplary embodiments, the central server may be configured to determine the location of an asset using one or more location determination algorithms, such as, but not limited to, triangulation methods, based on metadata associated with beacon signals received from the set of RF beacons. In some examples, the metadata associated with the beacon signals may include, but is not limited to, the signal strength of the beacon signals, the location of the RF beacons in the material handling environment, etc. As discussed, a set of RF beacons may include RF beacons that can receive beacon signals after one or more reflections. Such RF beacons may not be located in the LOS of the RF tag and may still receive beacon signals. Determining location based on beacon signals received from such RF beacons may be inaccurate and / or uncalibrated. Therefore, tracking assets within a material handling environment can be error-prone, potentially leading to further loss of productivity.

[0030] The implementation describes a material handling environment including one or more RF beacons installed at predetermined locations. A set of RF beacons can receive beacon signals from RF tags positioned on an asset. Furthermore, the set of RF beacons can transmit the beacon signals to a central server. The central server can be configured to determine the location of the asset within the material handling environment. For example, the asset may traverse the material handling environment, and the central server can be configured to track the asset's location based on beacon signals received from RF tags on the asset via the set of RF beacons. Since the set of RF beacons may be located within the direct LOS of the RF tags during the asset's traversal, the signal strength measured by the set of RF beacons can be accurate. Therefore, when the RF tags are located within the LOS of the set of RF beacons, the central server can determine the precise location of the asset within the material handling environment. The central server can be configured to store the LOS location of the asset and metadata associated with the beacon signals transmitted by the RF tags coupled to the asset.

[0031] When an asset approaches a storage shelf and / or aisle, a set of RF beacons can receive the beacon signal after multiple reflections (due to one or more surfaces approaching the RF tag coupled to the asset). Therefore, the location determined by the central server may be inaccurate and / or uncalibrated (hereinafter referred to as NLOS location). The central server may be configured to store NLOS locations and metadata associated with the beacon signals used to determine the NLOS location.

[0032] In some examples, the central server may further receive the correct location of the asset's NLOS position. In an exemplary implementation, the central server may receive the correct location from inertial sensors in an RF tag. Examples of inertial sensors may include, but are not limited to, accelerometers, gyroscopes, etc. In another example, the central server may receive the correct location manually from an operator moving the asset through a material handling environment. In yet another example, the central server may receive the correct location from a mobile computer 108 attached to the operator. For this purpose, the mobile computer 108 may be coupled to the operator's arm. Thus, when the operator moves the asset or places it on a shelf and / or aisle, the inertial sensors and / or GPS sensors on the mobile computer 108 can provide the correct location of the asset within the material handling environment. Similarly, when using machinery such as a forklift to move or traverse the asset, the inertial sensors and / or GPS sensors on the forklift can provide the accurate location of the asset for the NLOS position.

[0033] Subsequently, the central server can be configured to generate training data comprising one or more features and one or more labels. The one or more features correspond to the expected input of the ML model, and the one or more labels correspond to the expected output of the ML model. For this purpose, metadata associated with the beacon signal (used to determine the NLOS and LOS locations), the LOS location, and the NLOS location correspond to one or more features, and the exact location corresponds to one or more labels. The central server can then be further configured to train the ML model based on this training data.

[0034] A central server can be configured to use ML models to predict the accurate location of another asset within a material handling environment. For example, while another asset is moving through the material handling environment, the central server can be configured to track the asset's location based on beacon signals received from RF tags via a set of RF beacons. The central server can be configured to identify the LOS (Loss in State) and NLOS (Normally in State) location of the other asset. Furthermore, the central server can be configured to determine the metadata associated with the beacon signals used to determine the LOS and NLOS locations, respectively. Based on the LOS location, NLOS location, and / or the metadata associated with the beacon signals, the central server can be configured to predict the correct location corresponding to the NLOS location of the other asset.

[0035] Predicting the correct location using beacon signals received from the RF tag avoids the use of inertial and / or GPS sensors in the RF tag, machine, and / or mobile computer 108. Avoiding inertial and / or GPS sensors helps reduce the complexity of the material handling environment, thereby reducing the overall cost of operating in that environment. Furthermore, it significantly reduces the number of sensors that need to be deployed in the material handling environment. In addition, location tracking can be achieved simultaneously with tracking the contents stored in the RF tag.

[0036] Figure 1 An exemplary material handling environment 100 according to one or more embodiments described herein is illustrated. The material handling environment 100 includes a central server 102, one or more RF beacons 104, RF tags 106 coupled to assets 112, a mobile computer 108, a communication network 110, and a machine 114. The central server 102, one or more RF beacons 104, RF tags 106, mobile computer 108, and machine 114 are communicatively coupled to each other via the communication network 110.

[0037] Material handling environment 100 may correspond to a warehouse and / or any other space (such as an indoor environment) configured to store one or more assets 112. Additionally, material handling environment 100 may allow the assets 112 to be stored in one or more racks and / or aisles 116. Assets 112 may be placed in one or more racks and / or aisles 116. For example, an operator may manually place assets 112 on one or more racks and aisles 116. In another embodiment, machine 114 may be configured to place assets 112 on one or more racks and aisles 116.

[0038] Central server 102 may include suitable logic and / or circuitry that enables it to track assets 112 within material handling environment 100. In some examples, central server 102 receives beacon signals from RF tag 106 on asset 112 via a set of RF beacons (e.g., RF beacons 104a, 104b, and 104c) of one or more RF beacons 104. In some examples, the beacon signals correspond to radio frequency (RF) signals. Furthermore, central server 102 may be configured to determine the location of assets based on beacon signals, such as... Figure 3 In addition to the above, or alternatively, the central server 102 may be configured to classify the location of asset 112 into a first group of locations or a second group of locations. In some examples, the first group of locations may include locations determined when RF tag 106 is within the line of sight (LOS) of a group of RF beacons 104a and / or 104b. In some examples, the second group of locations may include locations determined when RF tag 106 is outside the LOS of a group of RF beacons 104a and / or 104b. Figure 3 The document further describes classifying the location of asset 112 into a first group of locations or a second group of locations. In an exemplary implementation, the central server 102 may be further configured to determine a precise set of locations within the second group of locations, such as... Figure 3 Further description follows. The central server 102 is further configured to train a machine learning (ML) model based on the first set of locations, the second set of locations, a set of accurate locations, and beacon signals, such as... Figure 3 Further description follows. Based on the ML model, the central server 102 can be configured to predict the location of another asset in the material handling environment 100, such as... Figure 5 Further description is provided below. The central server 102 can be implemented on any computing device without departing from the scope of this disclosure. (Combined with...) Figure 2 and Figure 3 The structure and operation of the central server 102 are further described.

[0039] One or more RF beacons 104 may include suitable logic and / or circuitry that enables the use of one or more known protocols, such as, but not limited to, Bluetooth.TM WiFi TM The system can transmit and / or receive data from RF tags 106 using RF protocols such as 3G, 4G, 5G, 2G, CDMA, CDMA2000, RFID, ZigBee, and / or any other RF-based communication protocols. One or more RF beacons 104 may include antenna elements that enable the transmission / reception of data from RF tags 106. One or more RF beacons 104 may be configured to receive beacon signals from RF tags 106. Furthermore, one or more RF beacons 104 may be configured to determine metadata associated with the beacon signals. Alternatively or additionally, one or more RF beacons 104 may be configured to transmit beacon signals to a central server 102.

[0040] RF tag 106 includes suitable logic and / or circuitry that enables RF tag 106 to use one or more known protocols, such as, but not limited to, Bluetooth. TM WiFi TM The RF tag 106 may periodically broadcast beacon signals using RF protocols such as 3G, 4G, 5G, 2G, CDMA, CDMA2000, RFID, ZigBee, and / or any other RF-based communication protocols. The RF tag 106 may further include a memory unit configured to store a unique identifier associated with the asset 112 to which the RF tag 106 is attached. In some examples, the RF tag 106 may be configured to broadcast the unique identifier on the beacon signal. The RF tag 106 may further include one or more inertial sensors (not shown), such as, but not limited to, accelerometers, gyroscopes, etc. The RF tag 106 may be configured to modify the periodicity of the broadcast beacon signal based on readings received from the one or more inertial sensors. If readings from the one or more inertial sensors indicate that the RF tag 106 is in motion, the RF tag 106 may be configured to increase the periodicity of the broadcast beacon signal. For example, the RF tag 106 may be configured to broadcast a beacon signal every 5 mms when the RF tag 106 is in motion (based on measurements from the one or more inertial sensors). If readings from one or more inertial sensors indicate that RF tag 106 is stationary, RF tag 106 can be configured to reduce the periodicity of the broadcast beacon signal. For example, RF tag 106 can be configured to broadcast a beacon signal every 1 second when RF tag 106 is stationary (based on measurements from one or more inertial sensors).

[0041] Mobile computer 108 includes suitable logic and / or circuitry that enables it to provide instructions to an operator to perform one or more tasks within material handling environment 100. In some examples, mobile computer 108 includes one or more image capture devices configured to scan barcodes printed on asset 112 and / or one or more barcodes on shelves and / or aisles 116. Mobile computer 108 may further include one or more inertial sensors and / or GPS sensors configured to determine location data (indicating the location of mobile computer 108 within material handling environment 100). Furthermore, mobile computer 108 may be configured to transmit location data to central server 102. Mobile computer 108 may correspond to any electronic device within material handling environment 100 that can be carried by an operator and is capable of capturing images.

[0042] Machine 114 may include one or more components that enable asset 112 to traverse within material handling environment 100. Machine 114 may include an engine unit, which may be electric and / or fuel-based. Additionally or alternatively, machine 114 may include one or more inertial sensors and / or GPS sensors that can be configured to generate location data. Furthermore, machine 114 may be configured to transmit location data to central server 102. Some examples of machine 114 may include, but are not limited to, conveyors, forklifts, etc.

[0043] Communication network 110 corresponds to the medium for the flow of content and messages between various devices in the material handling environment. Examples of communication network 110 may include, but are not limited to, Wi-Fi networks, wireless WANs, local area networks (LANs), or metropolitan area networks (MANs). Various devices in the material handling environment 100 can be connected to communication network 110 according to various wired and wireless communication protocols, such as Transmission Control Protocol and Internet Protocol (TCP / IP), User Datagram Protocol (UDP), and 2G, 3G, or 5G communication protocols.

[0044] Figure 2 A block diagram of a central server 102 according to one or more embodiments shown herein is illustrated. The central server 102 includes a processor 202, a memory device 204, an input / output device interface unit 206, a location determination unit 208, a training unit, and a prediction unit 212.

[0045] Processor 202 may be embodied as one or more microprocessors having a matching digital signal processor, one or more processors without a matching digital signal processor, one or more coprocessors, one or more multi-core processors, one or more controllers, processing circuitry, one or more computers, various other processing elements (including integrated circuits, such as application-specific integrated circuits (ASICs) or field-programmable gate arrays (FPGAs)) or some combination thereof.

[0046] Therefore, although in Figure 2 While illustrated as a single controller, in one exemplary embodiment, processor 202 may include multiple processors and signal processing modules. These multiple processors may be embodied in a single electronic device or distributed across multiple electronic devices that are collectively configured to serve as circuitry for central server 102. The multiple processors may operatively communicate with each other and may be collectively configured to perform one or more functions of the circuitry for central server 102, as described herein. In one exemplary embodiment, processor 202 may be configured to execute instructions stored in memory device 204 or otherwise accessible to processor 202. When these instructions are executed by processor 202, they may cause the circuitry for central server 102 to perform one or more functions, as described herein.

[0047] Regardless of whether processor 202 is configured by a hardware method, a firmware / software method, or a combination thereof, the processor may include an entity capable of performing operations and being configured accordingly according to embodiments of this disclosure. Thus, for example, when processor 202 is embodied as an ASIC, FPGA, etc., processor 202 may include hardware specifically configured to perform one or more of the operations described herein. Alternatively, as another example, when processor 202 is embodied as a runner of instructions (such as those that can be stored in memory device 204), these instructions may be specifically configured to configure processor 202 to perform one or more algorithms and operations described herein.

[0048] Therefore, the processor 202 used herein may refer to a programmable microprocessor, microcomputer, or one or more multiprocessor chips that can be configured by software instructions (applications) to perform various functions including those described in the various embodiments above. In some devices, multiple processors may be provided dedicated to wireless communication functions and one processor dedicated to running other applications. Software applications may be stored in internal memory before being accessed and loaded onto the processor. The processor may include sufficient internal memory to store application software instructions. In many devices, internal memory may be volatile or non-volatile memory such as flash memory or a combination of both. Memory may also be located within another computing resource (e.g., enabling computer-readable instructions to be downloaded via the Internet or another wired or wireless connection).

[0049] Memory device 204 may include suitable logic, circuitry, and / or interfaces adapted to store a set of instructions executable by processor 202 to perform predetermined operations. Some commonly known memory implementations include, but are not limited to, hard disks, random access memory, cache memory, read-only memory (ROM), erasable programmable read-only memory (EPROM) and electrically erasable programmable read-only memory (EEPROM), flash memory, magnetic tape cassettes, magnetic tape, disk storage devices or other magnetic storage devices, optical disc read-only memory (CD-ROM), digital versatile optical disc read-only memory (DVD-ROM), optical discs, circuitry configured to store information, or some combination thereof. In one exemplary embodiment, without departing from the scope of this disclosure, memory device 204 may be integrated with processor 202 on a single chip.

[0050] I / O device interface unit 206 may correspond to a communication interface that facilitates sending and receiving messages and data to and from various devices. Examples of I / O device interface unit 206 may include, but are not limited to, an antenna, an Ethernet port, a USB port, a serial port, or any other port suitable for receiving and transmitting data. I / O device interface unit 206 is adapted to various communication protocols, such as... Sending and receiving data and / or messages using Infra-Red, I2C, TCP / IP, UDP, and 2G, 3G, 4G, or 5G communication protocols.

[0051] The location determination unit 208 may include suitable logic and / or circuitry that enables it to receive beacon signals from the RF tag 106. Furthermore, the location determination unit 208 may be configured to receive metadata associated with the beacon signals, such as... Figure 3 Further description follows. Based on metadata associated with the beacon signal, the location determination unit 208 can be configured to determine the location of asset 112 within the material handling environment, such as... Figure 3 Further details are provided below. Additionally, the location determination unit 208 can be configured to generate time-series data on the location of asset 112, such as... Figure 3 Further description is provided below. Furthermore, the location determination unit 208 can be configured to classify locations in the time-series data of locations into a first group of locations and a second group of locations, such as... Figure 3 Further description is provided below. Additionally, the location determination unit 208 can be configured to receive location data from the mobile computer 108 and / or the machine 114, such as... Figure 3 Further description follows. Based on location data from mobile computer 108 and / or machine 114, location determination unit 208 can be configured to determine a third set of locations for the second set of locations, such as... Figure 3Further description is provided below. In an exemplary embodiment, the third set of positions corresponds to the calibrated / accurate positions of the second set of positions. The position determination unit 208 can be implemented using one or more known technologies, such as, but not limited to, field-programmable gate arrays (FPGAs) and / or application-specific integrated circuits (ASICs).

[0052] Training unit 210 may include suitable logic and / or circuitry that enables it to train an ML model based on time-series data of the location of asset 112 within the material handling environment 100, a third set of locations for a second set of locations, and metadata associated with beacon signals, such as... Figure 3 Further description is provided below. In an exemplary embodiment, training unit 210 may be configured to train an ML model using one or more known methods such as, but not limited to, logistic regression, Naive Bayes, convolutional neural networks (CNNs), etc. Training unit 210 may be implemented using one or more known techniques such as, but not limited to, field-programmable gate arrays (FPGAs) and / or application-specific integrated circuits (ASICs). Prediction unit 212 may include suitable logic and / or circuitry that enables prediction unit 212 to predict a fourth set of locations for another asset 112 in the material handling environment 100. In an exemplary embodiment, prediction unit 212 may be configured to use an ML model to predict a fourth set of locations for another asset 112 based on time-series data of the other asset's location and metadata associated with beacon signals (received from RF tags 106 associated with the other asset 112), such as... Figure 5 The prediction unit 212 can be implemented using one or more known techniques, such as, but not limited to, field-programmable gate arrays (FPGAs) and / or application-specific integrated circuits (ASICs).

[0053] Figure 3 A flowchart 300 is shown of a method for operating a central server 102 according to one or more embodiments described herein.

[0054] At step 302, the central server 102 may include means for receiving beacon signals from a set of RF beacons 104a and 104b, such as a processor 202, an I / O device interface unit 206, etc. As discussed, the RF tag 106 periodically broadcasts beacon signals that can be received by a set of RF beacons 104a and 104b that may be located within and / or outside the LOS of the RF tag 106.

[0055] Alternatively or in addition, processor 202 may be configured to determine metadata associated with beacon signals received from a set of RF beacons 104a and 104b. In an exemplary embodiment, the metadata associated with the beacon signals may include, but is not limited to, the signal strength of the beacon signal received by each of the RF beacons 104a and 104b, the unique ID of the RF beacon (through which the central server has received the beacon signal), the location of the RF beacon within the material handling environment 100, etc. Since processor 202 periodically receives beacon signals from each of the RF beacons 104a and 104b, processor may be configured to generate time-series data of the metadata associated with the beacon signals (received from each of the RF beacons 104a and 104b) for each of the RF beacons 104a and 104b. In the time-series data of the metadata associated with the beacon signals, the metadata associated with the beacon signals is timestamped based on the time when the RF beacon received the beacon signal. The following is an example of time-series data showing metadata associated with beacon signals:

[0056] Timestamp Unique ID Location of RF beacons Signal strength (in decibels) <![CDATA[T1]]> 1a (x1,y1,z1) 10Db <![CDATA[T2]]> 1a (x1,y1,z1) 3Db <![CDATA[T1]]> 1b (x2,y2,z2) 15Db <![CDATA[T1]]> 1c (x3,y3,z3) 20Db

[0057] Table 1: Time-series data of metadata associated with beacon signals

[0058] As can be observed from Table 1, at time T1, RF beacons 1a, RF beacons 1b, and RF beacons 1c receive beacon signals. Furthermore, the signal strengths of the beacon signals received by RF beacons 1a, RF beacons 1b, and RF beacons 1c can also be observed.

[0059] At step 304, the central server 102 may include means for receiving location data from one or more mobile computers (e.g., mobile computer 108) and / or one or more machines (e.g., machine 114) operating in the material handling environment 100, such as processor 202, I / O device interface unit 206, location determination unit 208, etc. In exemplary embodiments, location data may include, but is not limited to, accelerometer data, gyroscope data, and GPS data. As discussed, mobile computer 108 may include one or more of an inertial sensor (which also includes an accelerometer and a gyroscope) and a GPS sensor capable of generating location data. Similarly, machine 114 operating in the material handling environment 100 includes one or more inertial sensors and a GPS sensor capable of generating location data. Processor 202 may be configured to generate time-series data of the location data received from mobile computer 108 and / or machine 114 operating in the material handling environment 100. For this purpose, location determination unit 208 may periodically receive location data from mobile computer 108 and machine 114. In addition, location data indicates the location of mobile computers 108 and / or machines 114 within the material handling environment 100.

[0060] At step 306, the central server 102 may include means for determining the location of the asset 112 associated with the RF tag 106 based on metadata associated with beacon signals received from a set of RF beacons 104a and 104b, such as processor 202, I / O device interface unit 206, location determination unit 208, etc. In an exemplary embodiment, location determination unit 208 may be configured to determine the location of the RF tag 106 associated with the asset 112 using known methods. For example, location determination unit 208 may be configured to determine the location of the RF tag associated with the asset 112 using triangulation methods.

[0061] To this end, the location determination unit 208 may be configured to determine the location of each of a set of RF beacons 104a and 104b for which the processor 202 receives beacon signals. As discussed, the locations of one or more RF beacons 104 are predefined during deployment in the material handling environment 100. In an exemplary embodiment, the location determination unit 208 may be configured to determine the signal strength of the beacon signal received by each of the set of RF beacons 104a and 104b (based on time-series data of metadata associated with the beacon signal). Based on the signal strength of the beacon signal, the location determination unit 208 may be configured to estimate the distance between the RF tag 106 and the corresponding RF beacon (e.g., RF beacon 104a). In an exemplary embodiment, the location determination unit 208 may be configured to estimate the distance between the RF tag 106 and each of the set of RF beacons 104a and 104b using the inverse square relationship between distance and signal strength. Subsequently, the location determination unit 208 can be configured to define a virtual circle for each of the set of RF beacons 104a and 104b, where the estimated distance (between RF tag 106 and each of the set of RF beacons 104a and 104b) serves as the radius. Furthermore, the location determination unit 208 can be configured to determine the intersection point within the virtual circle created for the RF beacons 104a and 104b. In some examples, determining the intersection point may include determining the coordinates of the intersection point within the material handling environment 100. The location determination unit 208 can be configured to determine the coordinates of the intersection point based on the position of the set of RF beacons 104a and 104b within the material handling environment 100. In some examples, the location determination unit 208 can be configured to determine the coordinates of the intersection point using known geometric relationships. As discussed, the position of the set of RF beacons 104a and 104b can be predetermined during the deployment of the set of RF beacons 104a and 104b within the material handling environment 100. Based on the location of the intersection point within the material handling environment 100, the location determination unit 208 can be configured to estimate the location of the RF tag 106 within the material handling environment 100. The location of the RF tag 106 within the material handling environment 100 is considered as the location of the asset 112 within the material handling environment 100.

[0062] In an exemplary embodiment, the location determination unit 208 may be configured to determine the location of asset 112 based on time-series data of metadata from beacon signals (received via a set of RF beacons 104a and 104b). More specifically, the location determination unit 208 may be configured to determine the location of asset 112 at each moment the beacon signals are received. In some examples, the time-series data of asset 112's location may be used to monitor the location of asset 112 within the material handling environment 100. For example, the time-series data of asset 112's location may include locations indicating where asset 112 traverses the material handling environment 100. Furthermore, the time-series data of asset 112's location may include locations indicating the resting location of asset 112 (i.e., when asset 112 is stored in aisles and / or shelves 116 within the material handling environment 100).

[0063] At step 308, the central server 102 may include means for classifying the time-series location data of asset 112 into a set of transit locations and a set of stationary locations, such as processor 202, I / O device interface unit 206, location determination unit 208, etc. Combined with Figure 6 and Figure 7 The method for classifying the time-series data of the location of asset 112 is further described.

[0064] At step 310, the central server 102 may include means for classifying a set of traversal locations into a first set of locations or a second set of locations, such as processor 202, I / O device interface unit 206, location determination unit 208, etc. In an exemplary embodiment, the first set of locations may collectively indicate the traversal path of asset 112 through material handling environment 100. Furthermore, the location determination unit 208 may determine the first set of locations (within a set of traversal locations) when RF tag 106 on asset 112 is located in the LOS of a set of RF beacons 104a and 104b. Therefore, the location determination unit 208 may accurately determine the location within the first set of locations. Hereinafter, the first set of locations has been interchangeably referred to as a set of LOS locations.

[0065] In an exemplary implementation, when the RF tag 106 on asset 112 is outside the LOS of a set of RF beacons 104a and 104b (e.g., asset 112 may be approaching a channel that may cause reflections of the beacon signals), the location determination unit 208 may determine a second set of locations (out of a set of cross-locations). Therefore, the location determination unit 208 may not accurately determine the location within the second set of locations. Hereinafter, the second set of locations has been interchangeably referred to as a set of NLOS locations. Figure 8 The classification of a set of crossing locations is further described.

[0066] At step 312, the central server 102 may include means for determining a third set of positions relative to the second set of positions, such as processor 202, I / O device interface unit 206, position determination unit 208, etc. As discussed, the third set of positions corresponds to the calibration positions of the second set of positions. In an exemplary embodiment, position determination unit 208 may be configured to determine the third set of positions relative to the second set of positions based on position data received from a mobile computer 108 and / or machine 114 operating in the material handling environment 100. In another embodiment, position determination unit 208 may receive manual input from an operator, wherein the manual input relates to the third set of positions relative to the second set of positions. Figure 9 The determination of the third group of locations is further described.

[0067] At step 314, the central server 102 may include means, such as processor 202, I / O device interface unit 206, training unit 210, etc., for generating training data based on time-series data of the first set of locations, the second set of locations, the third set of locations, and beacon signals (received from RF tag 106 via a set of RF beacons 104a and 104b). In some examples, training unit 210 may be configured to define the first set of locations, the second set of locations, and the metadata associated with the beacon signals as one or more features in the training data. More specifically, the time-series data of the metadata associated with the beacon signals (included in the training data) may include only a portion of the time-series data of the metadata associated with the beacon signals, rather than the complete time-series data of the metadata associated with the beacon signals. In some examples, a portion of the time-series data may include metadata of the beacon signals used to determine the second set of locations. In another embodiment, the training data may include the complete time-series data of the metadata of the beacon signals used to determine the first set of locations and the second set of locations. As discussed, one or more features in the training data correspond to the expected inputs of the ML model (to be trained using the training data). Furthermore, training unit 210 can be configured to define a third set of positions (determined for the second set of positions) as one or more labels in the training data. As discussed, one or more labels correspond to the expected output of the ML model.

[0068] At step 316, the central server 102 may include means for training an ML model based on training data, such as processor 202, I / O device interface unit 206, training unit 210, etc. In an exemplary embodiment, training unit 210 may be configured to train the ML model using one or more known methods such as, but not limited to, logistic regression, convolutional neural networks, Keras, etc.

[0069] In some examples, the scope of this disclosure is not limited to training an ML model solely for assets 112 traversing a material handling environment. In an exemplary embodiment, training unit 210 may be configured to train another ML model for predicting the accurate location of assets 112 stored in aisles and / or shelves 116. As discussed, when assets 112 are stored in aisles and / or shelves 116, RF tags 106 may be located outside the LOS. Therefore, the location determined using metadata associated with beacon signals received from such RF tags 106 may be inaccurate. For this purpose, training unit 210 may train another ML model to enable the prediction of the accurate location of assets 112 stored in aisles and / or shelves 116. In an alternative embodiment, training unit 210 may be configured to further train the ML model to enable the ML model to predict the accurate location of assets 112 stored in aisles and / or shelves. Figure 4 One such method for further training the ML model is described.

[0070] Figure 4 A flowchart 400 illustrates a method for training another ML model capable of predicting the accurate location of asset 112 stored in shelves and / or aisles, according to one or more embodiments described herein.

[0071] At step 402, the central server 102 may include means for retrieving time-series data of metadata associated with beacon signals received from asset 112 (stored on shelves and / or in aisles), such as processor 202, I / O device interface unit 206, location determination unit 208, etc. More specifically, location determination unit 208 may be configured to retrieve time-series data of metadata associated with beacon signals for determining a set of stationary positions of asset 112 within material handling environment 100.

[0072] At step 404, the central server 102 may include means for retrieving a set of stationary locations of asset 112 (which are determined in step X08), such as processor 202, I / O device interface unit 206, location determination unit 208, etc.

[0073] At step 406, the central server 102 may include means for determining the precise location of asset 112 based on events, such as processor 202, I / O device interface unit 206, location determination unit 208, etc. In some examples, an event may include an operator using mobile computer 108 to scan a barcode printed on asset 112, and then scanning a barcode printed on the channel where asset 112 is placed and / or stored. After scanning the barcodes printed on asset 112 and channel 116, mobile computer 108 may be configured to retrieve barcode data from the barcodes printed on channel 116 and asset 112. The barcode data for channel 116 may include a unique identifier for channel 116. Thereafter, location determination unit 208 may be configured to retrieve the location of the channel from a lookup table that may include a mapping between the unique identifier of channel 116 and the corresponding location within the material handling environment 100. An exemplary lookup table is shown below, illustrating the mapping between the unique identifier of channel 116 and the corresponding location within the material handling environment 100:

[0074] Unique identifier for channel 116 Location in Material Handling Environment 100 Channel 1 (x4,y4,z4) Channel 2 (x5,y5,z5) Channel 3 (x6,y6,z6)

[0075] Table 2: A lookup table showing the mapping between the unique identifier of a channel and its corresponding location.

[0076] In some examples, the location determination unit 208 may be configured to regard the location of the channel (identified based on events) as the exact location of the asset 112.

[0077] At step 408, the central server 102 may include means for generating static asset training data based on metadata associated with beacon signals (for determining a set of stationary locations), a third set of locations of asset 112, and a set of stationary locations of asset 112 (determined using the beacon signal metadata), such as processor 202, I / O device interface unit 206, training unit 210, etc. More specifically, training unit 210 may be configured to define a set of stationary locations and beacon signal metadata as one or more features of the static asset training data. Furthermore, training unit 210 may be configured to define the precise location of asset 112 as one or more labels of the static asset training data.

[0078] At step 410, the central server 102 may include means for training another ML model based on static asset training data, such as processor 202, I / O device interface unit 206, training unit 210, etc. In an exemplary embodiment, training unit 210 may be configured to train another ML model using known techniques.

[0079] Figure 5 A flowchart 500 illustrates a method for predicting the location of a fourth set of assets within a material handling environment 100, according to one or more embodiments described herein.

[0080] At step 502, the central server 102 may include means for receiving beacon signals from another RF tag 106 associated with another asset via a set of RF beacons, such as processor 202, I / O device interface unit 206, location determination unit 208, etc. At step 504, the central server 102 may include means for determining the location of the other asset based on metadata associated with the beacon signals, as described in step 306, such as processor 202, I / O device interface unit 206, location determination unit 208, etc. Since processor 202 periodically receives beacon signals from the other RF tag, processor 202 generates time-series data of the beacon signal metadata, as described above in step 302; therefore, location determination unit 208 determines the time-series data of the location of the other asset 112.

[0081] At step 506, the central server 102 may include means for classifying time-series data of the location of another asset 112 into a set of cross-locations and a set of stationary locations, such as processor 202, I / O device interface unit 206, location determination unit 208, etc. In an exemplary embodiment, location determination unit 208 may be configured to classify the location of the other asset using the method described in step 308. At step 508, the central server 102 may include means for further classifying the set of cross-locations into a first set of locations or a second set of locations as described above in step 310, such as processor 202, I / O device interface unit 206, location determination unit 208, etc. As discussed, the first set of locations corresponds to a set of LOS locations, while the second set of locations corresponds to a set of NLOS locations.

[0082] At step 510, the central server 102 may include means for predicting a fourth set of locations for the second set of locations using an ML model, such as processor 202, I / O device interface unit 206, prediction unit 212, etc. In some examples, prediction unit 212 may be configured to provide the second set of locations and metadata associated with the beacon signal as input to the ML model. In response to this input, the ML model predicts the fourth set of locations for the second set of locations. Since the second set of locations is determined chronologically, prediction unit 212 can predict the fourth set of locations in the same chronological order as the second set of locations. Therefore, accurate locations within a set of accurate locations have associated timestamps. Furthermore, the timestamp associated with the accurate location (within a set of accurate locations) is the same as the timestamp of the corresponding location in the set of second locations.

[0083] At step 512, the central server 102 may include means for predicting the location of a channel based on a fourth set of locations predicted by an ML model, such as processor 202, I / O device interface unit 206, location determination unit 208, prediction unit 212, etc. Location determination unit 208 may be configured to retrieve an accurate location in the fourth set of locations based on a timestamp associated with each location in the fourth set of locations. For example, location determination unit 208 may be configured to retrieve an accurate location having the most recent timestamp in chronological order. Thereafter, based on this accurate location, location determination unit 208 may be configured to refer to a lookup table (which includes a mapping between channel identifiers and channel locations) to identify the channel closest to the accurate location. Thereafter, location determination unit 208 may be configured to consider the channel closest to the accurate location as a channel for placing or storing asset 112.

[0084] Therefore, the proposed implementation allows the central server to predict the location of the passageway for stored asset 112 without using additional sensors. For example, the proposed implementation allows the system to predict the passageway location (of stored asset 112) without scanning the barcode printed on passageway 116. Thus, the proposed implementation reduces the steps required by the operator and thereby increases the productivity of operations within the material handling environment 100.

[0085] Figure 6 A flowchart 600 is shown of a method for classifying time-series data of location according to one or more embodiments described herein.

[0086] At step 602, the central server 102 may include means for determining the duration elapsed between subsequent chronological moments of receiving a beacon signal via the RF beacon (in the RF beacon step), such as processor 202, I / O device interface unit 206, location determination unit 208, etc. As discussed, the RF tag may periodically transmit the beacon signal to the central server. Therefore, processor 202 may be configured to determine the duration elapsed between subsequent chronological moments of receiving a beacon signal via the RF beacon (in the RF beacon step). In another embodiment, processor 202 may be configured to determine the duration elapsed between subsequent chronological moments of receiving a beacon signal from one of a set of RF beacons 104a and 104b.

[0087] At step 604, the central server 102 may include means for determining whether the elapsed duration is less than a duration threshold, such as processor 202, I / O device interface unit 206, location determination unit 208, etc. In an exemplary embodiment, the duration threshold may be predetermined based on the periodicity of beacon signal transmission from RF tag 106. As discussed, RF tag 106 transmits beacon signals more frequently when it is in motion compared to when RF tag 106 is stationary. The motion of RF tag 106 is detected based on readings from inertial sensors in RF tag 106.

[0088] If the location determination unit 208 determines that the elapsed duration is less than a duration threshold, then the location determination unit 208 may be configured to execute step 606. However, if the location determination unit 208 determines that the elapsed duration is greater than a duration threshold, then the location determination unit 208 may be configured to execute step 608.

[0089] At step 606, the central server 102 may include means for classifying the location of asset 112 into a set of cross-locations, such as processor 202, I / O device interface unit 206, location determination unit 208, etc., to determine the location of the asset using metadata associated with a beacon signal received at a later time in chronological order.

[0090] At step 608, the central server 102 may include means for classifying the location of asset 112 into a set of stationary locations, such as processor 202, I / O device interface unit 206, location determination unit 208, etc., to determine the location of the asset using metadata associated with a beacon signal received at a later time in chronological order.

[0091] In some examples, the scope of this disclosure is not limited to classifying time-series data of the location of asset 112 based on the duration elapsed between receiving beacon signals. In exemplary embodiments, location determination unit 208 may be configured to classify time-series data based on events. In some examples, an event may include an operator scanning a barcode printed on asset 112 and then scanning a barcode printed on the channel where asset 112 is placed and / or stored. Figure 7 One such method is described in [the document / article].

[0092] Figure 7 A flowchart 700 is shown of a method for classifying time-series data of location according to one or more embodiments described herein.

[0093] At step 702, the central server 102 may include means for receiving barcode data associated with asset 112 and barcode data associated with channel 116 where asset 112 has been stored by an operator, such as processor 202, I / O device interface unit 206, location determination unit 208, etc. In some examples, processor 202 may be configured to receive barcode data associated with asset 112 and barcode data associated with channel 116. In some examples, the operator may place asset 112 on channel 116. Subsequently, in order to indicate to the central server 102 that the task of placing asset 112 is complete and to record the location of the placed asset 112, the operator may scan the barcode printed on asset 112 and then scan the barcode printed on channel 116. Mobile computer 108 may send the barcode data associated with asset 112 and channel 116 to central server 102. Furthermore, central server 102 may associate the location of asset 112 with channel 116 based on the barcode data associated with asset and channel.

[0094] At step 704, the central server 102 may include means for detecting an event in response to receiving barcode data relating to the location of the passage and barcode data relating to the asset 112, such as processor 202, I / O device interface unit 206, location determination unit 208, etc. At step 706, the central server 102 may include means for retrieving time-series data of the location of the asset 112 determined during a predetermined duration prior to the event, such as processor 202, I / O device interface unit 206, location determination unit 208, etc. At step 708, the central server 102 may include means for determining the locations in the time-series data of the asset 112's location determined during the predetermined duration prior to the event as a set of crossing locations, such as processor 202, I / O device interface unit 206, location determination unit 208, etc. For this purpose, it is assumed that an operator can cause the asset 112 to cross the material handling environment 100 during a predetermined duration prior to the event. Therefore, the location of the asset 112 during the predetermined duration is considered a set of crossing locations.

[0095] At step 710, the central server 102 may include means for determining a set of stationary positions after an event, such as processor 202, I / O device interface unit 206, position determination unit 208, etc.

[0096] In another embodiment, the central server 102 may receive data from one or more inertial sensors in the RF tag 106, as well as beacon signals. Based on the accelerometer data, the position determination unit 208 may be configured to determine whether the RF tag 106 is in motion. Therefore, the position determination unit 208 may be configured to classify the position (determined using metadata from the received beacon signals) into a set of crossing positions or a set of stationary positions.

[0097] Figure 8 A flowchart 800 is shown of a method for classifying a set of crossing locations into a first group of locations and a second group of locations according to one or more embodiments described herein.

[0098] At step 802, the central server 102 may include means for retrieving a first set of traversal locations from a set of traversal locations in chronological order, such as a processor 202, an I / O device interface unit 206, a location determination unit 208, etc. In some examples, the first set of traversal locations may include locations determined during a first time period within a predetermined duration.

[0099] At step 804, the central server 102 may include means for determining the distance between each pair of time-sequentially determined locations in the first set of traversed locations, such as a processor 202, an I / O device interface unit 206, a location determination unit 208, etc. For example, the first set of traversed locations includes locations L1, L2, L3, and L4 determined at times T1, T2, T3, and T4, respectively. For this purpose, the location determination unit 208 may be configured to determine the distance between L1 and L2, the distance between L2 and L3, and the distance between L3 and L4.

[0100] At step 806, the central server 102 may include means for determining whether the distance determined for each pair of time-ordered locations is within a distance threshold, such as processor 202, I / O device interface unit 206, location determination unit 208, etc. If location determination unit 208 determines that the distance between a pair of time-ordered distances is greater than the distance threshold, then location determination unit 208 may be configured to perform step 808. However, if location determination unit 208 determines that the distance between each pair of time-ordered distances is within the distance threshold, then location determination unit 208 may be configured to perform step 810.

[0101] At step 808, the central server 102 may include means for determining a position in the first set of traversed positions as a second set of positions, such as a processor 202, an I / O device interface unit 206, a position determination unit 208, etc. At step 810, the central server 102 may include means for determining a position in the first set of traversed positions as a first set of positions, such as a processor 202, an I / O device interface unit 206, a position determination unit 208, etc.

[0102] Figure 9 A flowchart 900 is shown of a method for determining a third set of positions for a second set of positions, according to one or more embodiments described herein.

[0103] At step 902, the central server 102 may include means for identifying events, such as processor 202, I / O device interface unit 206, location determination unit 208, etc. As discussed, the event corresponds to receiving barcode data associated with asset 112 and / or the channel through which asset 112 is stored. At step 904, the central server 102 may include means for identifying mobile computer 108 from which barcode data is received, such as processor 202, I / O device interface unit 206, location determination unit 208, etc. In some examples, processor 202 may be configured to identify mobile computer 108 based on metadata associated with data packets (which include barcode data) received by the central server. In some examples, the data packets may include detailed information related to the MAC address and / or IP address of mobile computer 108. Processor 202 may be configured to consider mobile computer 108 as associated with asset 112.

[0104] At step 906, the central server 102 may include means for identifying a third set of locations from time-series data of location data received from the identified mobile computer 108, such as processor 202, I / O device interface unit 206, location determination unit 208, etc. The location determination unit 208 may determine the third set of locations based on timestamps associated with the time-series data of the location data of the mobile computer 108 and timestamps associated with the second set of locations. For example, the location determination unit 208 may be configured to retrieve the third set of locations from time-series data of location data (received from the mobile computer 108) having timestamps identical to those of the second set of locations. In another embodiment, the location determination unit 208 may be configured to retrieve the third set of locations from time-series data of location data (received from the mobile computer 108) having timestamps within a predefined time range of the timestamps of the second set of locations. The predetermined time range may be defined during the configuration of the central server.

[0105] At step 908, the central server 102 may include means for determining the third set of locations as a third set of locations relative to the second set of locations, such as processor 202, I / O device interface unit 206, location determination unit 208, etc.

[0106] In some examples, the scope of this disclosure is not limited to using mobile computer 108 to determine a third set of locations relative to the second set of locations. In an exemplary embodiment, location determination unit 208 may be configured to determine a third set of locations relative to the second set of locations using the location of machine 114 through which asset 112 traverses. As discussed, machine 114 includes inertial sensors and / or GPS sensors for generating location data for machine 114. Figure 9 The method described herein allows the location determination unit to be configured to retrieve a third set of locations from time-series data of the location data of machine 114. Furthermore, as discussed, the third set of locations corresponds to the exact locations of the second set of locations.

[0107] Figure 10 An exemplary scenario 1000 of training an ML model according to one or more embodiments described herein is shown.

[0108] Exemplary scenario 1000 illustrates time-series data of the location 1002 of asset 112 determined based on metadata (shown by 1004) of beacon signals received by a central server via a set of RF beacons. It can be observed that the time-series data of the location 1002 of asset 112 includes a first set of locations 1006 and a second set of locations 1008. For this purpose, the second set of locations 1008 has one or more clusters 1008a, 1008b, and 1008c of locations because the beacon signals are reflected by one or more surfaces near asset 112. Therefore, exemplary scenario 1000 shows that the second set of locations includes uncalibrated locations. Additionally, exemplary scenario 1000 illustrates a time-series (shown by 1010) of location data obtained from mobile computer 108 and / or machine 114. As discussed, the location data obtained from mobile computer 108 and / or machine 114 includes a third set of locations for the second set of locations.

[0109] In an exemplary implementation, training unit 210 may be configured to train ML model 1012 using a set of accurate locations, a second set of locations, and metadata associated with beacon signals.

[0110] The above detailed description has illustrated various implementations of the device and / or process using block diagrams, flowcharts, schematic diagrams, examples, and illustrations. Each function and / or operation within such block diagrams, flowcharts, schematic diagrams, or illustrations can be implemented individually and / or collectively by a wide variety of hardware, provided that such block diagrams, flowcharts, schematic diagrams, or illustrations contain one or more functions and / or operations.

[0111] It should be noted that each block in the flowchart, and combinations thereof, can be implemented by various means, such as hardware, firmware, circuitry, and / or other devices associated with the execution of software including one or more computer program instructions. For example, one or more processes in the process described above can be embodied by computer program instructions, which can be stored in the non-transitory memory of a device employing embodiments of the present disclosure and executed by the processor of that device. These computer program instructions can instruct a computer or another programmable device to operate in a particular manner such that the instructions stored in a computer-readable storage memory produce an article of writing whose execution implements the function specified in the flowchart block.

[0112] The embodiments of this disclosure can be configured as methods, mobile devices, backend network devices, etc. Therefore, embodiments can include various means, including entirely hardware or any combination of software and hardware. Furthermore, embodiments can take the form of a computer program product on at least one non-transitory computer-readable storage medium, the computer program product having computer-readable program instructions (e.g., computer software) embodied in the storage medium. Similarly, embodiments can take the form of computer program code stored on at least one non-transitory computer-readable storage medium. Any suitable computer-readable storage medium can be used, including non-transitory hard disks, CD-ROMs, flash memory, optical storage devices, or magnetic storage devices.

[0113] In one implementation, the examples of this disclosure may be implemented via an application-specific integrated circuit (ASIC). However, the implementations disclosed herein may be equivalently implemented, in whole or in part, on a standard integrated circuit as one or more computer programs (e.g., one or more programs running on one or more computer systems), one or more programs running on one or more processing circuits (e.g., microprocessor circuits), one or more programs running on one or more processors (e.g., microprocessors), firmware, or virtually any combination thereof.

[0114] Furthermore, those skilled in the art will recognize that the exemplary mechanisms disclosed herein may be distributed as program products in a variety of tangible forms, and the exemplary embodiments are equally applicable regardless of the specific type of tangible instruction-bearing medium used to actually perform the distribution. Examples of tangible instruction-bearing media include, but are not limited to, the following: recordable media such as floppy disks, hard disk drives, CD-ROMs, digital magnetic tapes, flash drives, and computer memory.

Claims

1. A method for tracking assets, the method comprising: The processor determines one or more locations of the asset within the indoor environment based on metadata associated with RF signals received from RF tags linked to the asset. The processor identifies a first set of locations among the one or more locations, wherein the first set of locations corresponds to the calibration location of the asset where the RF tag on the asset is located within the line of sight (LOS) of an RF beacon installed in the indoor environment; The processor identifies a second set of locations from the one or more locations, wherein the second set of locations corresponds to an uncalibrated location of the asset where the RF tag on the asset is located outside the LOS of the RF beacon within the indoor environment; The processor determines a third set of locations for the asset based on the correlation between a second timestamp associated with the second set of locations and a first timestamp associated with time-series data of location data received from one of the mobile computers or machines. The third set of locations is a calibration location of the second set of locations, wherein the third set of locations is identified from the time-series data of the location data, such that the first timestamp associated with the third set of locations is the same as the second timestamp associated with the second set of locations. as well as The processor trains a machine learning (ML) model based on the first set of locations, the second set of locations, the third set of locations, and the metadata associated with the RF signal, wherein the ML model is configured to predict a fourth set of locations of another asset within the indoor environment.

2. The method of claim 1, further comprising the processor determining one or more locations of another asset within the indoor environment.

3. The method of claim 2, wherein one or more locations of the other asset include a first set of locations corresponding to a calibrated location of the other asset in the indoor environment, and a second set of locations, wherein the second set of locations includes an uncalibrated location of the other asset in the indoor environment.

4. The method of claim 3, wherein the fourth set of positions corresponds to the calibration position of the other asset of the second set of positions of the other asset.

5. The method of claim 1, further comprising the processor classifying one or more locations of the asset into a set of crossing locations or a set of stationary locations based on the periodicity of the RF signals received from the RF tag.

6. The method of claim 5, wherein the first set of locations and the second set of locations of the asset are identified from a set of crossing locations of the asset.

7. The method of claim 1, further comprising receiving accelerometer data from the RF tag by the processor.

8. The method of claim 7, the method further comprising the processor classifying one or more locations of the asset into a set of crossing locations or a set of stationary locations based on the accelerometer data.

9. The method of claim 1, the method further comprising receiving input corresponding to scanning a barcode on the asset.

10. The method of claim 9, further comprising determining one or more locations of the asset based on the metadata associated with the RF signal received from the RF tag on the asset during a predetermined time period prior to the first moment.

11. A central server for tracking assets, the central server comprising: processor; A storage device communicatively coupled to the processor, the storage device including a set of instructions executable by the processor for: Based on metadata associated with RF signals received from RF tags associated with the asset, determine one or more locations of the asset within the indoor environment; Identify a first set of locations among the one or more locations, wherein the first set of locations corresponds to the calibration location of the asset within the indoor environment, wherein the RF tag on the asset is within the line of sight (LOS) of an RF beacon installed within the indoor environment; Identify a second set of locations from the one or more locations, wherein the second set of locations corresponds to locations where RF tags on assets within the indoor environment are not within the line of sight of RF beacons, and wherein the second set of locations corresponds to uncalibrated locations of RF tags within the indoor environment; A third set of locations of the asset within an indoor environment is determined based on the correlation between timestamps associated with the second set of locations and timestamps associated with time-series data of location data received from one of the mobile computers or machines, wherein the third set of locations corresponds to calibrated locations of the second set of locations, and wherein a first timestamp associated with the second set of locations and a second timestamp associated with the time-series data of the location data are the same; and Based on the first set of locations, the second set of locations, the third set of locations, and metadata associated with the machine learning (ML) model, an ML model is trained, wherein the ML model is configured to predict a fourth set of locations of the asset within an indoor environment when another RF tag on another asset leaves the LOS of the RF beacon.

12. The central server of claim 11, wherein the processor is further configured to determine one or more locations of another asset within an indoor environment.

13. The central server according to claim 12, wherein, The other asset's one or more locations include a first set of locations corresponding to the other asset's calibrated locations in an indoor environment, and a second set of locations, wherein the second set of locations includes the other asset's uncalibrated locations in an indoor environment.

14. The central server according to claim 13, wherein, The fourth set of positions corresponds to the calibration position of the other asset in the second set of positions of the other asset.

15. The central server according to claim 11, wherein, The processor is further configured to classify one or more locations of the asset into a set of crossing locations or a set of stationary locations based on the periodicity of the RF signals received from the RF tag.

16. The central server according to claim 15, wherein, The first set of locations and the second set of locations of the asset are identified from the first set of crossing locations of the asset.

17. The central server according to claim 16, wherein, The processor is further configured to classify one or more locations of the asset into a set of crossing locations or a set of stationary locations based on accelerometer data received from the RF tag.

18. The central server of claim 11, wherein the processor is further configured to receive input corresponding to scanning a barcode on the asset.

19. The central server according to claim 18, wherein, The processor is further configured to determine one or more locations of the asset based on the metadata associated with the RF signals received from the RF tag on the asset during a predetermined time period prior to the first moment.

20. A method for tracking assets, the method comprising: The processor determines one or more locations of the asset within an indoor environment based on metadata associated with RF signals received from an RF tag associated with the asset. The one or more locations include a first set of locations and a second set of locations. The first set of locations corresponds to a calibrated location of the asset where the RF tag on the asset is within the line of sight (LOS) of an RF beacon installed within the indoor environment. The second set of locations corresponds to an uncalibrated location of the asset where the RF tag on the asset is outside the LOS of the RF beacon installed within the indoor environment. The processor, using a machine learning (ML) model, predicts a fourth set of locations based on the first set of locations, the second set of locations, and a third set of locations determined based on location data received from a mobile computer. The third set of locations corresponds to calibrated locations of the second set of locations, and a first timestamp associated with the second set of locations is the same as a second timestamp associated with time-series data of the location data received from one of the mobile computer and the machine. The location of the channel for the stored assets is predicted based on the fourth set of locations.

Images